Endocrinology Volume 1, Issue 1
AS PRESENTED IN THE ROUNDS
R OUNDS ™
OF THE DIVISION
ENDOCRINOLOGY AND METABOLISM,
Combination therapy with oral
hypoglycemic agents: Leading with Innovation
Serving with Compassion
Rationale and principles ST. MICHAEL’S HOSPITAL
A teaching hospital afﬁliated with the University of Toronto
A M I R K . H A N N A , M B, B C H , F R C P C , F A C P
The objective of this inaugural issue of Endocrinology Rounds is to review the impact
of intensive and early glucose control on the complications of diabetes, understand the
characteristics of the different classes of oral hypoglycemic agents, highlight the beneﬁts
Members of the Division of
of combination therapy with oral hypoglycemic agents, and provide an overview of recent
Endocrinology and Metabolism
combination therapy studies.
at St. Michael’s Hospital
L. LEITER, MD (HEAD)
Studies in people with type 1 and 2 diabetes1-3 have demonstrated that intensive treat-
EDITOR, ENDOCRINOLOGY ROUNDS
ment of hyperglycemia results in signiﬁcant reduction in microvascular complications and a
tendency for decreased macrovascular disease. Epidemiological analyses of the United G. BOOTH, MD
Kingdom Prospective Diabetes Study (UKPDS)4 found a continuous relationship between
P. CONNELLY, PHD
glucose levels and both micro- and macrovascular disease.
The DCCT/EDIC (Diabetes Control and Complications Trial/Epidemiology of Dia- C. DERZKO, MD
betes Interventions and Complications Research Group) study 5 followed up most of the J. GOGUEN, MD
original cohort of the DCCT for 4 years after the original results were analyzed. The group
A. HANNA, MD
initially assigned to intensive treatment continued to show signiﬁcantly less microvascular
complications when compared to the conventional treatment group, even though the D. JENKINS, MD, PHD
difference in HbA1c values between the 2 groups became much smaller. In the Kumamoto R. JOSSE, MB
8-year update study, differences in HbA1c between the intensive and conventional groups,
T. MURRAY, MD
seen at the end of the ﬁrst 6 years of the study, were maintained. Continuing protection
from microvascular complications was noted.6 D. NG, MD
Management protocols derived from The Clinical Practice Guidelines for the Manage- R. PATTEN, MD
ment of Diabetes in Canada7 recommend a step-wise approach to glucose control in people
L. RAO, PHD
with type 2 diabetes. This approach commences with life-style changes and is followed by
oral agent monotherapy, combinations of oral agents, and ﬁnally, adding or substituting W. SINGER, MD
insulin. Moving from one step of treatment to the next is recommended when target levels R. VOLPE, MD
of blood glucose are not attained after a waiting period of 2 to 4 months. This approach,
V. VUKSAN, PHD
however, accepts treatment failure and tolerates the detrimental effects of glucose toxicity
on insulin secretion and sensitivity. T. WOLEVER, MD, PHD
A new paradigm of early utilization of combination therapy has, therefore, been sug- M. WOO, M.D, PHD
gested. It is aimed at attaining ideal blood glucose levels early on in the course of therapy.
This approach is followed by efforts to maintain these ideal levels with frequent measure-
ments of pre- and postprandial glucose levels, HbA1c. Target levels of these variables are
based on those found in people without diabetes (eg, a HbA1c level that is < 6%). This St. Michael’s Hospital
approach has the following beneﬁts: 6121-61 Queen St. E.
• it targets the different pathogenetic factors contributing to hyperglycemia, (ie, insulin Toronto, Ont. M5C 2T2
Fax: (416) 867-3696
resistance, impaired beta cell function, and increased hepatic glucose production);
• it uses sub-maximal doses of different oral agents, decreasing the possible side effects of The editorial content of Endocrinology
such medications; Rounds is determined solely by the
Division of Endocrinology and Metabolism,
• it avoids the effects of glucose toxicity. St. Michael’s Hospital, University of Toronto.
There are two studies – Action to Control Cardio- Insulin sensitizers
vascular Disease in Diabetes (ACCORD) and the VA
The thiazolidinediones: pioglitazone and rosiglitazone
Diabetes Trial 8,9 – that are now evaluating the effect of
attaining normal glycemia (HbA1c < 6%), among other Pioglitazone and rosiglitazone are members of the
interventions, on the occurrence of macrovascular com- thiazolidinedione family, a group of insulin sensitizing
plications in patients with type 2 diabetes. agents that enhance sensitivity to insulin in adipose
tissue, striated muscle and, to a lesser extent, in the liver.
Characteristics of the different classes of They decrease hepatic glucose production and increase
oral hypoglycemic agents insulin-mediated glucose uptake. The mechanism of
Insulin secretagogues action is not fully elucidated. Both agents are known to
Insulin secretagogues act by binding to the different bind to the nuclear receptor PPAR–γ (peroxisome prolif-
sulfonylurea receptor sites on the plasma membrane of erator-activated receptor–gamma), which regulates the
beta cells in the pancreatic islets. The resulting closure of transcription of several genes involved in insulin-medi-
the KATP channels leads to membrane depolarization, ated glucose uptake in peripheral tissues, bringing about
opening of the calcium channels, and inﬂux of calcium, increased glucose transporter translocation (GLUT- 4) to
resulting in insulin secretion. This group of drugs is the cell membrane. Their effect appears to be achieved
utilized in an attempt to correct the quantitative (overall primarily by direct action on fat cells. Both drugs,
decreased insulin secretion in response to hyperglycemia) therefore, amplify insulin action and reduce insulin
and qualitative (absent first phase insulin secretion) resistance.18-20
abnormalities of insulin secretion in type 2 diabetes. The glucose-lowering effect of these drugs is similar
Sulfonylureas increase insulin secretion; however, to the sulphonylureas. In patients naïve to oral agents,
only some members of this class partially restore ﬁrst the decrease in HbA1c ranges between 0.8% and 1.9%,
phase insulin excursion. This group of drugs is usually depending on the dose used. In a group of patients with
well tolerated, effective in lowering blood glucose, type 2 diabetes previously treated with other oral hypo-
but has a 1%-10% yearly secondary failure rate. Hypo- glycemic agents, the decrease in HbA1c ranged between
glycemia and weight gain are the main side effects, the 0.3 and 0.9%.21-25 Thiazolidinediones are slow to reach
magnitude of which vary among different members of their maximum therapeutic effect, taking between 8-12
this class. Gliclazide and glimepiride tend to be associated weeks to do so. In a study that compared rosiglitazone
with less hypoglycemia and weight gain.10,11 with pioglitazone in patients with type 2 diabetes previ-
ously treated with troglitazone, there was no difference in
Meglitinide analogues glycemic control as measured by HbA1c.24 Other charac-
These drugs represent a new family of insulin secreta- teristics of these drugs are shown in Table 1.
gogues. Structurally, they are non-sulfonylurea agents.
They are absorbed rapidly, stimulate insulin release Side effects of pioglitazone and rosiglitazone
within a few minutes, have a more potent effect on Side effects occur in a small number of patients and
enhancing ﬁrst phase insulin secretion, are rapidly metab- have resulted in <1% withdrawal from clinical studies.
olized in the liver, and are excreted mainly in bile. The Edema was reported in about 5%-15% of patients treated
meglitinide analogues are short-acting. They appear to with pioglitazone in monotherapy and with different
lower postprandial glucose levels more than the sulfonyl- combinations.39 Edema tended to be more severe in
ureas and are associated with less hypoglycemia. Glucose combination with insulin and sulfonylureas as compared
control, as measured by HbA1c, is equivalent to the to monotherapy. About 1%-6% of patients on rosiglita-
sulfonylureas and metformin. These drugs are therefore zone experience edema.24,64 The differences in edema
more suitable for patients with type 2 diabetes who have incidence could be related to design of the studies men-
a variable lifestyle and elevated postprandial glucose tioned. In patients with history of CHF or NYHA class
levels.12-16 Nateglinide is a phenylalanine derivative. In 3 and 4, the risk of heart failure is such that these drugs
a study in a group of non-diabetic volunteers that should not be used. Decreased hemoglobin of 3 to
compared nateglinide 120 mg to repaglinide 2 mg taken 10 gm/L was noted in some studies.24,39 An average 0.7-
10 min before food, nateglinide resulted in a more rapid 3.5 kg weight gain has been noted, related to both
and short-lived stimulation of insulin secretion, resulting increased fat mass and ﬂuid retention. A higher degree of
in lower meal-related glucose excursions.17 Repaglinide is weight gain was noted when these drugs were combined
available in Canada, while nateglinide, which is available with insulin.24,33 In long-term use, no significant hepatic
in the United States and other countries, is not yet avail- toxicity has been observed.
able in Canada. The recommended dose of repaglinide
is 0.5 to 4 mg, up to 4 times a day; nateglinide is recom- Recommendations for pioglitazone and rosiglitazone use
mended in a dosage of 120 mg tid, taken shortly before Both drugs are recommended in patients with type 2
meals. diabetes who are not adequately controlled or unable to
Table 1: Characteristics of the thiazolidinediones • Patients with a history of congestive heart failure,
(pioglitazone, rosiglitazone) or in the NYHA class 3 and 4 should not be given
• There is evidence of preservation of islet cell insulin
• Pregnancy: fertility may be restored in patients with
content and improved beta cell function.26,28
polycystic ovary syndrome (PCO) as insulin resist-
• Fat distribution studies suggest a decrease in visceral
fat in the context of overall increased fat mass related to
ance decreases. Precautions to prevent pregnancy
increased peripheral fat. There was a signiﬁcant reduc- should be discussed with the patient and docu-
tion in waist/hip ratio in a 26-week study with rosiglita- mented until further experience is gained with
zone.24,29,30 these drugs.
• A slight decrease (2.3 mm Hg) in diastolic blood pres-
sure was noted with rosiglitazone compared to baseline, Metformin
and in systolic blood pressure with pioglitazone com-
Metformin acts on the liver by reducing hepatic
pared to acarbose (10 mm Hg decrease with pioglita-
zone, compared to 1.5 mm Hg decrease with acar- glucose production. It suppresses gluconeogenesis
bose).31,32 mainly by potentiating the effects of insulin, reducing
• There is reduction in free fatty acid levels24 and a favor- hepatic extraction of certain substrates (eg, lactate), and
able effect on lipid parameters with both pioglitazone opposing the effects of glucagon.41 Insulin-stimulated
and rosiglitazone, reducing the LDL/HDL ratio. In studies glucose uptake into skeletal muscles is enhanced mainly
with pioglitazone, there was a tendency towards more as a result of increased movement of glucose trans-
reduction in triglyceride levels as compared to rosiglita-
porters into the cell membrane.42 Metformin suppresses
zone.24,33 However, there are no head to head com-
parisons between these two drugs, making a deﬁnitive fatty acid oxidation and reduces hypertriglyceridemia.
conclusion hard to ascertain. A meta-analysis of 9 randomized controlled trials
• There is no evidence of clinically signiﬁcant drug interac- compared metformin with sulfonylurea. Glucose was
tions with other oral hypoglycemic agents, oral contra- reduced by 2 mmol/L and HbA1c by 0.9% compared to
ceptives, nifedipine, digoxin, ranitidine, or war farin.33,34 placebo, with no significant weight change. These
• Studies with rosiglitazone show that its pharmaco- results were similar to the sulfonylureas, except in terms
kinetics are not altered in renally impaired patients or of body weight (an increase of 1.7 kg with sulfony-
those on hemodialysis; as well, the pharmacokinetics lureas, compared to a 1.2 kg weight loss in metformin-
of pioglitazone are not altered in moderate to severe
renally impaired patients. Thus, there is no need to
change the drug dosage in patients with impaired renal
function.33-37 Alpha-glucosidase inhibitors
• Preliminary studies with rosiglitazone show a tendency The alpha-glucosidase inhibitors inhibit the effects of
towards decreased albumin excretion rate and intestinal enzymes responsible for carbohydrate absorp-
albumin/creatinine ratio.24,38 This, however, has not yet tion. Alpha-glucosidases are enzymes located on the
been veriﬁed in long-term controlled studies.
brush border of the small intestine where they break
down oligosaccharides and disaccharides into mono-
tolerate their current oral hypoglycemic therapy. In saccharides, which are then absorbed in the proximal
patients in whom other oral hypoglycemic therapies are jejunum. They do not directly affect beta cells and do
contraindicated, the thiazolidinediones provide an effec- not cause hypoglycemia. They attenuate postprandial
tive and safe alternative. I feel that the present data40 glycemia. There is minimal absorption of alpha-glucosi-
justify the addition of a thiazolidinedione in patients who dase inhibitor from the gut, about 2%-3%; the rest remains
are not meeting treatment targets while on maximum enteric.44 Both acarbose and miglitol reduce glucose less
doses of combination sulfonylurea and metformin. It is effectively than glyburide (0.75% compared to 1%) and
also my opinion that in certain patients who require large cause more gastrointestinal side-effects (ﬂatulence seen
amounts of insulin – especially if they are not attaining twice as much in the treatment group versus the control
glycemic targets – the addition of a thiazolidinedione groups). They do not cause weight gain or hypo-
could improve glucose parameters and reduce insulin glycemia.45 Acarbose is available in Canada, while miglitol
dosage. This combination, however, should be used with is not.
extreme caution because of the possibility of accelerated Both acarbose and miglitol are given at 50-100 mg
edema, weight gain, and congestive heart failure. The tid, taken with the first bite of the meal. Starting with a
recommended doses for pioglitazone are 15, 30, or 45 mg low dose (25 mg once daily), and gradually increasing
once daily; for rosiglitazone, it is 4 mg once or twice per the dose, seems to reduce the gastrointestinal side effects.
day or 8 mg once daily.
Contraindications Primary combination therapy
• Abnormal liver function tests: AST > 2.5 times In a study by Herman et al, 56 newly diagnosed
normal is a contraindication to starting patients on patients with type 2 diabetes were randomized to either
this group of drugs to metformin or glyburide monotherapy, or a combination.
Follow-up for 6 months showed a decrease in fasting nateglinide and metformin groups, respectively.
plasma glucose (FPG) by 3.2 mmol/L and in HbA1c Fasting plasma glucose was reduced by 2.4 versus
by 1.5% in all groups. Weight increased by 3.1 kg in 1.6 and 2.4 mmol/L, respectively. The area under the
the glyburide group, but stayed constant in the others.46 curve of glucose excursion was signiﬁcantly lower
A second study by Haupt et al,47 combining with both nateglinide and combination therapy com-
metformin with a maximum dose of sulfonylurea, pared to placebo and metformin.50
in patients who failed monotherapy, resulted in a
decrease in fasting plasma glucose by 2.8 mmo/L, Repaglinide + metformin
postprandial glucose by 4.1 mmol/L, and HbA1c by 83 patients with type 2 diabetes who failed to
1.9%.47 achieve treatment targets while on treatment with
In a third study,48 806 patients with newly diag- metformin (HbA1c >7.1%) were randomized to
nosed type 2 diabetes who did not attain optimal metformin (n=27), repaglinide (n=29), or a combi-
control with life-style changes were randomized to nation of both (n=27). The same metformin dose
placebo, glyburide, metformin, and two ﬁxed combi- was continued in patients receiving mono or com-
nations of metformin/glyburide (M/G): 250/1.25 mg bination therapy; however, repaglinide was adjusted
and 500/2.5 mg. Final doses for each active treatment over a 4-8 week titration period. Patients were
group were: 5.3 mg glyburide; 1317 mg metformin, followed for a period of 3 months after reaching
and 557/2.8 mg and 818/4.1 mg M/G, respectively. the final repaglinide dose. Fasting plasma glucose
Decrease in HbA1c was signiﬁcantly greater with the decreased by 2.2 mmol/L (P<0.0003) and HbA1c
ﬁxed combinations (1.48% and 1.53%, respectively), by 1.4% in the combination therapy group
compared with placebo (0.21%), (P<0.001 for both); (P< 0.0016); 60% of the patients receiving combi-
glyburide: 1.24%, (P<0.02 for M/G 250/1.25, and nation therapy reached HbA1c <7% compared to
P<0.005 for M/G 500/2.5) and metformin: 1.03% only 20% in either monotherapy groups.51
(P<0.001 for both) at week 20. Sixty-ﬁve to 70% of
patients on the ﬁxed combinations reached HbA1c Other combination studies with repaglinide
< 7%, signiﬁcantly more than other treatment arms. Combination studies done with metformin,
troglitazone, rosiglitazone, pioglitazone and insulin
Combination therapy with oral agents showed a reduction in HbA1c by 1.3% to 1.7%.55-59
after failure with monotherapy Missed-lunch studies have shown less hypoglycemia
Glyburide + metformin when compared to glyburide
In a study by DeFronzo et al, 632 patients were
treated with diet, or diet + a sulfonylurea. Open label Rosiglitazone and pioglitazone
glyburide was stopped. Patients were randomized to: combination studies
glyburide (n=209), metformin (n=210), or both medi- Studies combining rosiglitazone or pioglitazone
cations (n=213). Fasting plasma glucose decreased by with sulphonylureas, metformin, and insulin showed
0.9 mmol/L in the glyburide group, 0.4 mmol/L in additional glucose lowering. Fasting plasma glucose
the metformin group, and by 3.4 mmol/L in the and HbA1c showed signiﬁcant decline. HbA1c
group receiving the combination of both drugs declined by 1.6% to 1.7% compared to placebo.55-59
(P<0.001). HbA1c decreased by 0.2% in the gly- A recent study of triple therapy using troglitazone in
buride group, 0.4% in the metformin group, and patients who fail to attain optimal glycemic control
1.7% in the combination therapy group (P<0.001). with maximum doses of metformin and glyburide,
Body weight remained the same with glyburide, showed an average decrease in HbA1c of 1.4% com-
decreased by 3.8 kg with metformin, and by 0.4 kg pared to placebo.40 At the present time, the glita-
with the combination.49 zones are not currently indicated for use in combina-
tion with insulin in Canada.
Nateglinide + metformin
701 patients with type 2 diabetes were random- Alpha-glucosidase inhibitors
ized to nateglinide 120 mg tid alone, metformin Combination therapy of acarbose 60,61 and migli-
500 mg tid alone, or a combination of nateglinide tol with sulfonylureas, metformin, and insulin
120 mg and metformin 500 mg tid after a washout results in a reduction in HbA1c of 0.5% to 1.4%.
period. They were followed for 24 weeks. The results
showed that HbA1c and FPG decreased more with Conclusion
combination therapy as compared to monotherapy Glycemic control in patients with type 2 dia-
(P<0.0001). HbA1c was reduced by 1.4% in the combi- betes should aim at normalization of glucose para-
nation group compared to 0.5% and 0.8% in the meters: fasting, postprandial glucose and HbA1c.
9. Duckworth WC, McCarren M, Abraira C, Veterans Affairs Diabetes
Table 2: Rationale for targeting postprandial Trial (VADT), Glucose control and cardiovascular complications: the
glucose levels VA Diabetes Trial. Diabetes Care 2001;24(5):942-945.
10. Tessier D, Dawson K, Tetrault JP, et al. Glibenclamide vs gliclazide
in type 2 diabetes of the elderly. Diabet Med 1994;11(10):974-80.
• Postprandial glucose levels correlate better with
11. Holstein P, et al. Glimepiride less hypoglycemia than glyburide.
HbA1c values Diabetologia 2000;43:A40.
• There is a correlation between postprandial glucose 12. Malaisse WJ. Stimulation of insulin release by non-sulfonylurea
hypoglycemic agents: the meglitinide family. Horm Metab Res 1995;
and cardiovascular disease (The DECODE study 63). 27:263-266.
• The availability of interventions targeting post- 13. Wolffenbuttel BHR, Landgraf R, on behalf of the Dutch and
German Repaglinide Study Group. A 1-year multi-center random-
prandial glucose control: dietary interventions (high ized double-blind comparison of repaglinide and glyburide for the
soluble ﬁber foods, foods with low glycemic index), treatment of type 2 diabetes. Diabetes Care 1999;22:463-467.
insulin secretagogues of the meglitinide group, and 14. Nateglinide: A structurally novel insulin secretion agent. Drugs of
rapid-acting insulin analogues Today 2001;37(suppl F)1-16.
15. Damsbo P, Clauson P, Marbury TC, et al. A double-blind randomized
comparison of meal-related glycemic control by repaglinide and gly-
buride in well controlled type 2 diabetic patients. Diabetes Care
Early introduction of combination therapy with oral 16. Kristensen JS, Frandsen KB, et al. The frequency of severe hypo-
glycemia is reduced with repaglinide treatment compared with
agents is recommended, rather than the stepwise sulphonylurea treatment Eur J Endocrinol 1999;140(Suppl 1)
approach presently advocated in the Clinical Prac- 19:Abstract 57.
17. Kalbag, JB, Walter,YH, Nedelman, JR, et al. Mealtime glucose regu-
tice Guidelines. Target levels of control should be lation with nateglinide in healthy volunteers: comparison with
the same as those found in people without diabetes. repaglinide and placebo. Diabetes Care 2001;24(1):73-77.
18. Lehmann JM, Moore LB, Smith-Oliver TA, et al. An antidiabetic thi-
Failure of combination oral agents to attain such azolidinedione is a high afﬁnity ligand for peroxisome proliferator-
targets is an indication for additional therapy or activated receptor. J Biol Chem 1995;270:12953-12956
19. Saltiel AR, Olefsky JM. Thiazolidinediones in the treatment of
substitution of insulin. Postprandial glucose levels insulin resistance and type II diabetes. Diabetes 1996;45:1661-1669.
correlate with HbA1c and cardiovascular disease 20. Willson TM, Brown PJ, Sternbach DD, et al. The PPARs: from
orphan receptors to drug discovery. J Med Chem 2000; 43:527-550.
better than fasting glucose levels (Table 2). Targeting 21. Aronoff S, Rosenblatt S, Braithwaite S, et al. Pioglitazone hydrochlo-
postprandial glucose levels is made possible by ride monotherapy improves glycemic control in the treatment of
patients with type 2 diabetes. Diabetes Care 2000;23:1605-1611.
interventions that improve first phase insulin secre- 22. Raskin P, Rappaport EB, Cole ST, et al. Rosiglitazone short-term
tion, delay carbohydrate absorption, or provide monotherapy lowers fasting and postprandial glucose in patients
with type II diabetes. Diabetologia 2000;43:278-284.
rapid acting insulins. 23. Nolan JJ, Jones NP, Patwardhan R, et al. Rosiglitazone taken once
daily provides effective glycemic control in patients with Type 2 dia-
betes mellitus. Diabet Med 2000;17:287-294.
Amir K. Hanna, MB, BCh, FRCPC, FACP, is an
24. Lebovitz HE, Dole J, Patwardhan R, et al. Rosiglitazone monothera-
Associate Professor, Department of Medicine, py is effective in patients with type 2 diabetes. J Clin Endocrinol
University of Toronto, and Director, Diabetes Clinic,
25. Khan MA, St Peter JV, Neafus KL, et al. A prospective, randomized
St. Michael’s Hospital, Toronto, Ontario comparison of the metabolic effects of pioglitazone in patients with
type 2 diabetes who were previously treated with troglitazone.
References 26. Matthews DR, Bakst A, Weston WM, et al. Rosiglitazone decreases
1. The Diabetes Control and Complications Trial Research Group. The insulin resistance and improves beta-cell function in patients with
effect of intensive therapy of diabetes on the development and pro- type 2 diabetes. Diabetologia 1999;42(Suppl 1): Abstract 858.
gression of long-term complications in insulin-dependent diabetes 27. Rosenstock J, for the pioglitazone HCI study group. Improved
mellitus. N Engl J Med 1993;329: 977-986. insulin sensitivity and beta cell response with pioglitazone therapy
2. Ohkubo Y, Kishikawa, H, Araki E, et al. Intensive insulin therapy pre- suggested by HOMA analysis. Diabetologia 2000; 43(Suppl 1):A192.
vents the progression of diabetic microvascular complications in 28. Jones NP, et al. Rosiglitazone reduces plasma insulin and its precur-
Japanese patients with non-insulin dependent diabetes mellitus: a sors while decreasing glycemia in type 2 diabetes. Diabetes 1999;48
randomized prospective 6-year study. Diabetes Res Clin Pract 1995; (Suppl 1):Abstract 859.
28:103-117. 29. Kelly IE, Han TS, Walsh K, et al. Effects of a thiazolidinedione com-
3. UKPDS Group. Intensive blood glucose control with sulphonylureas pound on body fat and fat distribution of patients with type 2 dia-
or insulin compared with conventional treatment and risk of betes. Diabetes Care 1999;22:288-293.
complications in patients with type 2 diabetes (UKPDS 33). Lancet 30. Miyazaki,Y, Mahankali A, Matsuda M, et al. Effect of pioglitazone on
1998;352:837-853. abdominal fat distribution and insulin sensitivity in patients with
4. Stratton IM, Adler AL, Neil HA, et al. Association of glycaemia with type 2 diabetes mellitus (T2DM). Diabetes 2000;49(Suppl.1):A299.
macrovascular and microvascular complications of Type 2 Diabetes 31. St John Sutton M, Dole J, Rappaport EB. Rosiglitazone does not
(UKPDS 35): prospective observational study. BMJ 2000;321:405- adversely affect cardiac structure or function in patients with type 2
412. diabetes. Diabetes 1999;42(Suppl 1):A102.
5. Retinopathy and nephropathy in patients with type 1 diabetes four 32. Scherbaum W, Goke B, for the German Pioglitazone Study Group,
years after a trial of intensive therapy. The Diabetes Control and Dusseldorf and Munich, Germany. Pioglitazone reduces blood pres-
Complications Trial/ Epidemiology of Diabetes Interventions and sure in patients with type 2 diabetes mellitus. Diabetes 2001;50
Complications Research Group. N Engl J Med 2000;342(6):381- (Suppl 2):A462.
389. 33. Avandia (rosiglitazone maleate) Product Monograph, CPS.
6. Shichiri M, Kishikawa H, Ohkubo Y, Wake N. Long-term results of 34. Crijns-Kortboyer J, Eckland D. Pioglitazone has low potential for
the Kumamoto study on optimal diabetes control in diabetic drug interaction. Diabetologia 1999;42(Suppl 1):A228.
patients. Diabetes Care 2000;23(suppl 2):B21-9. 35. Chapelsky MC, Thompson K, Miller A. Effect of renal impairment
7. Meltzer S, Leiter L, Daneman D, et al. 1998 Clinical Practice on the pharmacokinetics of rosiglitazone (RSG). Clin Pharmacol
Guidelines for the Management of Diabetes in Canada. CMAJ 1998; Ther 1999;65:185.
159 Suppl 8:S1-29. 36. Thompson K, Zussman B, Miller A. Pharmacokinetics of rosiglita-
8. Action to Control Cardiovascular Disease in Diabetes (ACCORD) zone are unaltered in hemodialysis patients. Clin Pharmacol Ther
Study Protocol 2. 1999;65:186.